Dr Dahmani is worried that the favorable impact of ketamine described in our analyses may only apply to adults but not to children. He refers to previously published, apparently similar, analyses that were restricted to the pediatric population.
We agree with Dr Dahmani that in general, results found in adult trials may not necessarily be applicable to a pediatric population. We could have excluded pediatric trials from our analyses to yield a more homogenous patient population with data from adults only. However, as there is no obvious biological basis as to why ketamine should have a different effect in children aged 6 to 16 years compared with adults, we decided to include all relevant trials independent of patients' age. Eventually, none of the trials reporting on opioid consumption at 24 hours postoperatively included into our analyses was performed in children. Also, only 1 of the 9 trials reporting on pain intensity at 24 hours included children. In a sensitivity analysis, we found that results on pain intensity did not change significantly when data from children were excluded. Finally, we found no evidence of a difference in the effect of ketamine on postoperative pain intensity at 24 hours in children compared with adults.
To make his point, Dr Dahmani cites 2 meta-analyses that combined very heterogenous data from trials testing in children the analgesic efficacy of ketamine administrated by a variety of routes. In the analysis by Michelet et al.,2 for instance, none of the 11 included studies examined the effect of ketamine as an adjuvant to an opioid-PCA. Also, trial sequential analyses regarding the outcome “opioid consumption” suggested that the Michelet meta-analysis was underpowered.
Eligibility criteria applied in our meta-analysis were strict. We considered only randomised trials comparing patients receiving an intravenous opioid-based PCA regimen with ketamine (experimental group) with patients receiving the same opioid-PCA regimen but without ketamine (control group). Trials where ketamine was administered by an additional, alternative route were not considered.
In conclusion, as our meta-analysis included only 2 middle-sized pediatric studies with a cumulative number of 109 children, we agree with Dr Dahmani that the results of our analyses should be applied with caution to the pediatric population. Consequently, the research agenda should not ignore the need for randomized studies testing the usefulness of ketamine as an adjuvant to opioid-PCA in children.